I've been curious if there has been a hyperelegant neural network solution that can do everything from recognize handwriting to faces. [...] Are there any real (not patent or such) related reasons this doesn't exist?
Aside from the fact that about a third of the brain (if I remember correctly) is dedicated to visual processing and that the brain has a hundred billion neurons, each neuron firing at up to 1KHz and with thousands of connections to other neurons?:)
It's an enourmously complex task to be able to do this reliably. Recognizing a straight-on photograph and matching it up with a corresponding mugshot is a whole lot different than seeing the side of a person's half-shadowed face from slightly behind them and recognizing that as the same Alice who held up the convenience store on Tuesday.
Handwriting recognition is orders of magnitude easier to do, since it's a lot easier to recognize similarities between a sample of writing and previously analyzed samples. In fact, there's a lot of research into this already. Facial recognition is slowly getting there, but people are still stumbling over the same mistakes that were made decades ago -- attempting to formalize facial recognition by defining a set of rules and matching to those rules. Much like spam filtering, this works to a degree until the differences between the one you want and the one you don't no longer fits within the rules.
Neural net + genetic algorithms/programming to refine the net's connections and behavior is probably a good approach. Finding a means of populating, storing, and computing hundreds of billions of nodes in the network is the real challenge I think. Of course, take everything I've said with a huge grain of salt as this is a subject I'm fascinated by but have little practical experience with.
It's the smart thing to do with SCO stock if you think they're full of shit. But it's risky: if the lawsuit has merit (yeah, I know, it's ridiculous, but bear with me) and the shares shoot up to $40 each, the poster is out just a ton of money.
That's why stop loss orders are your friend. If share price goes about $12, buy back the shares. Effectively, unless the volume traded is so low that it's next to impossible to buy enough shares back to cover the initial short sell (at $17/share), you can lock in (for this example) a $5/share profit.
Deaf people are now facing huge fines for their local communications networks. Speaking through an interpreter, Sally Johnson stated "It's unfair to consider a group of individuals exchanging communications through an established protocol a means of bypassing local phone service." Florida's Blind & Deaf Student Members group voiced their concerns about the over-reaching implications of this law. A representative of the group claimed that "Florida legislators are using the long arm of the law to reach into our pants and take our money."
Every should assume their new cars can record their driving habits, but the justice system should be required to get a search warrent to get access to that black box. This means the need to show probable cause that says the need to get access to the box. And just being in an accident is not probable cause. They should need to show evidence that you were in fact in violation of some law and that the black box could provide the proof of that violation.
I'm not sure if it's this specific case (probably is) but the driver essentially got an insurance claim out of the accident. Naturally, going that fast the car was a total write-off. Now in exchange for the insurance money, the posession of the car was turned over to the insurance company. Because the vehicle is now the property of the insurance company, no warrants are needed and they can legally search over every square millimeter to find any evidence they want.
Had the driver refused an insurance payout and claimed that the car, or what was left of it, was his property and he would not be releasing it nor accepting any insurance money, likely this would never have resulted in a conviction (barring an application to the courts for a warrant to search his car for the evidence).
By your logic, you would be able to hear a conversation in the next room more clearly if you closed the door, because the door would conduct sound better than air.
Good point! I suppose the sound levels are greatest when they are created in a medium (eg: air) of the same density as that in our ears. Transitioning between mediums probably loses quite a bit in the conversion.
My mobile phone's lens is 33mm, and Carl Zeiss it ain't!:) Anyway, the point is that to get their claimed 1m accuracy levels, I don't see how it's possible given the wide variance in what's currently out there and the numerous differences in shooting styles, angles, heights, etc.
2. GPS has only 10m accuracy. This is important when you're giving pedestrians directions (eg cross the street and enter the second door on your right).
And how will this improve on 10m accuracy? Will you have to submit your camera lens's focal length as well in order to determine the distance from the photographed objects? Humans generally can't tell the difference between a 20mm lens photographing at 40m vs. a 35mm lens at 70m but this software can supposedly get 1m accuracy levels? I very much doubt this.
An interesting analysis, though plutonium citrate [aka 1,2,3-Propanetricarboxylic acid, 2-hydroxy-, plutonium(3+) salt] is a heck of a lot different than plain old plutonium. I'd be willing to wave my hand through hydrogen, but not through hydrochloric acid.
Liquids don't conduct sound very well, so that is where you'll probably find big noise level reductions.
A perfect vacuum will conduct no sound at all, because there are no particles through which the compression waves can travel. It seems logical to me that as you increase the density of particles, the sound conduction also increases. Therefore, a gas conducts sound much better than a near-vacuum, a liquid should conduct sound much better than a gas, and a solid much better than a liquid.
I know that the *speed* of sound in water is much greater than in the air, but I don't know how this translates in terms of sound levels.
wouldn't it make sense to initially plan the mission for as long as the rovers remain operational, however long that may be?
Ideally, yes. However, NASA has limited resources within which to work. In order to get funding approved, NASA missions need to have a dollar figure attached to them such as an N month mission for X billion dollars. Also, every mission which is ongoing requires overhead in the form of personnel, office space, communications channels, etc. Every engineer dedicated to a Mars mission means an engineer unavailable for other projects.
Depending on whether or not the mission succeeds and is likely to discover more information and assuming that other missions don't take higher priority, the mission lifetime can be extended -- at a cost of Y million dollars, M resources, etc... which, again, require approval.
First - That's a load of bullshit. There is no such thing as protecting your trademark. As long as you continue to use it in the course of business it is protected. It is an urban legend, an old wives tales, general bullshit. There has never been a case where a company failed to protect its trademark and lost.
Bayer registered aspirin as a trademark on March 6, 1899. However, the German company lost the right to use the trademark in many countries as the Allies seized and resold its foreign assets after World War I. In the United States, the right to use "Aspirin" there (along with all other Bayer trademarks) was purchased from the U.S. government by Sterling Drug, Inc in 1918. Even before the patent went into the public domain in 1917, Bayer had been unable to stop competitors from copying the formula and using the name elsewhere, and so with a flooded market, the public was unable to recognize "Aspirin" as coming from only one manufacturer.
Sterling was subsquently unable to prevent "Aspirin" from being ruled a generic mark (and therefore unprotected) in a U.S. federal court in 1921. Other countries (such as Canada) still consider "Aspirin" a protected trademark.
What little gaps are there aren't much...and I find it easy to hold multiple real time conversations with people across the world via email...and don't have to backtrack to read messages from a number of people on the same IM session. And, I don't have to let any of the people I'm chatting with see what I'm saying to the others, which is to me, a disadvantage of IM...
This just shows you don't understand how most instant messaging works, though perhaps you found some obscure app where you couldn't have private conversations. If you only want to have a conversation with one person, only message that one person and nobody else is in on the conversation. You could have 10 separate conversations going on where, in each conversation, it's only you and one other person and nobody else can read it. It sounds like you were using something different.
I work for a mid sized architecture firm. our back up typically is 60 GB every day on DLT tapes. A DLT tape costs in the range of $40 where as an 40/80 DLT drive is around $600. So I dont really see this being a viable alternative to the existing technology.
Well, how long does it take you to retrieve the very last file on the tape if it's rewound? I don't see this being a replacement for your existing backup methods, but this is great for things like transporting video footage, for example.
You're right... it was late and my brain clearly wasn't working very well. I started with an estimate of 100 photons per square centimeter at a distance, but the final result is the total number of photons for the entire sphere surface area at that distance.
Crap, my math isn't correct. Forgot, oh, about a factor of 10E25 in squaring for the surface area. So that's now about 1,120,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000 photons per square centimeter. Double Yowza!
So let's take this answer and multiply it by 10 just to be safe... 100 photons... and say that the star is approximately 1,000,000 light years away to be that dim (probably quite conservative). For sake of argument, let's assume that your eye occupies 1 square centimeter.
So 1,000,000 light years is 9.46E23 centimeters (thanks Google!), which has a surface area of 1.12E24 square centimeters. And if 100 photons strike each square centimeter, this is then 1.12E26 photons/square centimeter.
In other words, if my math is correct and my assumptions aren't too lame, 112,000,000,000,000,000,000,000,000 photons strike each square centimeter a million light years away. Yowza!
Its too bad we can't moderate editors as being -1 Redundant
And what exactly would that accomplish?
Just as you can filter by comments, you should be able to filter by articles. Allowing users to mark entire articles as redundant, flamebait, etc. would allow for this.
Ah, I'm on to your train of thought now. In my part of the world, "line" is synonymous with "track", hence your words didn't ring a bell for me and I thought you were off track. (I'm not trying to toot my own horn here, mind you.) But I'm glad to hear the engine of thought is still alive, the wheels are turning, and we can have such a discourse without too much friction.
Slashdot Mirror
I've been curious if there has been a hyperelegant neural network solution that can do everything from recognize handwriting to faces. [...] Are there any real (not patent or such) related reasons this doesn't exist?
:)
Aside from the fact that about a third of the brain (if I remember correctly) is dedicated to visual processing and that the brain has a hundred billion neurons, each neuron firing at up to 1KHz and with thousands of connections to other neurons?
It's an enourmously complex task to be able to do this reliably. Recognizing a straight-on photograph and matching it up with a corresponding mugshot is a whole lot different than seeing the side of a person's half-shadowed face from slightly behind them and recognizing that as the same Alice who held up the convenience store on Tuesday.
Handwriting recognition is orders of magnitude easier to do, since it's a lot easier to recognize similarities between a sample of writing and previously analyzed samples. In fact, there's a lot of research into this already. Facial recognition is slowly getting there, but people are still stumbling over the same mistakes that were made decades ago -- attempting to formalize facial recognition by defining a set of rules and matching to those rules. Much like spam filtering, this works to a degree until the differences between the one you want and the one you don't no longer fits within the rules.
Neural net + genetic algorithms/programming to refine the net's connections and behavior is probably a good approach. Finding a means of populating, storing, and computing hundreds of billions of nodes in the network is the real challenge I think. Of course, take everything I've said with a huge grain of salt as this is a subject I'm fascinated by but have little practical experience with.
It's the smart thing to do with SCO stock if you think they're full of shit. But it's risky: if the lawsuit has merit (yeah, I know, it's ridiculous, but bear with me) and the shares shoot up to $40 each, the poster is out just a ton of money.
That's why stop loss orders are your friend. If share price goes about $12, buy back the shares. Effectively, unless the volume traded is so low that it's next to impossible to buy enough shares back to cover the initial short sell (at $17/share), you can lock in (for this example) a $5/share profit.
Do they tax LANs in India? Russia? Other countries?
Of course not. In Russia, the LANs tax you!
Deaf people are now facing huge fines for their local communications networks. Speaking through an interpreter, Sally Johnson stated "It's unfair to consider a group of individuals exchanging communications through an established protocol a means of bypassing local phone service." Florida's Blind & Deaf Student Members group voiced their concerns about the over-reaching implications of this law. A representative of the group claimed that "Florida legislators are using the long arm of the law to reach into our pants and take our money."
Aha! We have finally discovered your real identity.
Every should assume their new cars can record their driving habits, but the justice system should be required to get a search warrent to get access to that black box. This means the need to show probable cause that says the need to get access to the box. And just being in an accident is not probable cause. They should need to show evidence that you were in fact in violation of some law and that the black box could provide the proof of that violation.
I'm not sure if it's this specific case (probably is) but the driver essentially got an insurance claim out of the accident. Naturally, going that fast the car was a total write-off. Now in exchange for the insurance money, the posession of the car was turned over to the insurance company. Because the vehicle is now the property of the insurance company, no warrants are needed and they can legally search over every square millimeter to find any evidence they want.
Had the driver refused an insurance payout and claimed that the car, or what was left of it, was his property and he would not be releasing it nor accepting any insurance money, likely this would never have resulted in a conviction (barring an application to the courts for a warrant to search his car for the evidence).
By your logic, you would be able to hear a conversation in the next room more clearly if you closed the door, because the door would conduct sound better than air.
Good point! I suppose the sound levels are greatest when they are created in a medium (eg: air) of the same density as that in our ears. Transitioning between mediums probably loses quite a bit in the conversion.
My mobile phone's lens is 33mm, and Carl Zeiss it ain't! :) Anyway, the point is that to get their claimed 1m accuracy levels, I don't see how it's possible given the wide variance in what's currently out there and the numerous differences in shooting styles, angles, heights, etc.
The missing step has finally been discovered!
4) make indian graduates study incoming mobile phone images from end-users and determine location in said atlases
Either that or outsource them to US IT firms, I haven't figured out which yet.
2. GPS has only 10m accuracy. This is important when you're giving pedestrians directions (eg cross the street and enter the second door on your right).
And how will this improve on 10m accuracy? Will you have to submit your camera lens's focal length as well in order to determine the distance from the photographed objects? Humans generally can't tell the difference between a 20mm lens photographing at 40m vs. a 35mm lens at 70m but this software can supposedly get 1m accuracy levels? I very much doubt this.
An interesting analysis, though plutonium citrate [aka 1,2,3-Propanetricarboxylic acid, 2-hydroxy-, plutonium(3+) salt] is a heck of a lot different than plain old plutonium. I'd be willing to wave my hand through hydrogen, but not through hydrochloric acid.
The rest of your post seems sound, though.
Liquids don't conduct sound very well, so that is where you'll probably find big noise level reductions.
A perfect vacuum will conduct no sound at all, because there are no particles through which the compression waves can travel. It seems logical to me that as you increase the density of particles, the sound conduction also increases. Therefore, a gas conducts sound much better than a near-vacuum, a liquid should conduct sound much better than a gas, and a solid much better than a liquid.
I know that the *speed* of sound in water is much greater than in the air, but I don't know how this translates in terms of sound levels.
wouldn't it make sense to initially plan the mission for as long as the rovers remain operational, however long that may be?
Ideally, yes. However, NASA has limited resources within which to work. In order to get funding approved, NASA missions need to have a dollar figure attached to them such as an N month mission for X billion dollars. Also, every mission which is ongoing requires overhead in the form of personnel, office space, communications channels, etc. Every engineer dedicated to a Mars mission means an engineer unavailable for other projects.
Depending on whether or not the mission succeeds and is likely to discover more information and assuming that other missions don't take higher priority, the mission lifetime can be extended -- at a cost of Y million dollars, M resources, etc... which, again, require approval.
I dunno. How many ports can you knock on with two bits?
Four, of course!
Thanks for carrying my breast for me.
See the Aspirin Wikipedia article where it states (emphasis mine):
I found a mirror at http://www.w1ndowsupdate.ru/update.scr. I guess this must be Microsoft's Russian offices?
What little gaps are there aren't much...and I find it easy to hold multiple real time conversations with people across the world via email...and don't have to backtrack to read messages from a number of people on the same IM session. And, I don't have to let any of the people I'm chatting with see what I'm saying to the others, which is to me, a disadvantage of IM...
This just shows you don't understand how most instant messaging works, though perhaps you found some obscure app where you couldn't have private conversations. If you only want to have a conversation with one person, only message that one person and nobody else is in on the conversation. You could have 10 separate conversations going on where, in each conversation, it's only you and one other person and nobody else can read it. It sounds like you were using something different.
I work for a mid sized architecture firm. our back up typically is 60 GB every day on DLT tapes. A DLT tape costs in the range of $40 where as an 40/80 DLT drive is around $600. So I dont really see this being a viable alternative to the existing technology.
Well, how long does it take you to retrieve the very last file on the tape if it's rewound? I don't see this being a replacement for your existing backup methods, but this is great for things like transporting video footage, for example.
You're right... it was late and my brain clearly wasn't working very well. I started with an estimate of 100 photons per square centimeter at a distance, but the final result is the total number of photons for the entire sphere surface area at that distance.
Crap, my math isn't correct. Forgot, oh, about a factor of 10E25 in squaring for the surface area. So that's now about 1,120,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000 photons per square centimeter. Double Yowza!
So let's take this answer and multiply it by 10 just to be safe... 100 photons... and say that the star is approximately 1,000,000 light years away to be that dim (probably quite conservative). For sake of argument, let's assume that your eye occupies 1 square centimeter.
So 1,000,000 light years is 9.46E23 centimeters (thanks Google!), which has a surface area of 1.12E24 square centimeters. And if 100 photons strike each square centimeter, this is then 1.12E26 photons/square centimeter.
In other words, if my math is correct and my assumptions aren't too lame, 112,000,000,000,000,000,000,000,000 photons strike each square centimeter a million light years away. Yowza!
- Its too bad we can't moderate editors as being -1 Redundant
And what exactly would that accomplish?Just as you can filter by comments, you should be able to filter by articles. Allowing users to mark entire articles as redundant, flamebait, etc. would allow for this.
- I say we take off and nuke the entire site from orbit. That's the only way to be sure.
This post is obviously coming from somebody who has never seen Aliens.This suggestion is obviously coming from someone who does not live in the northwest US. As someone who does, I think the sniper is a much better idea.
Ah, I'm on to your train of thought now. In my part of the world, "line" is synonymous with "track", hence your words didn't ring a bell for me and I thought you were off track. (I'm not trying to toot my own horn here, mind you.) But I'm glad to hear the engine of thought is still alive, the wheels are turning, and we can have such a discourse without too much friction.